AI tool comparison
Apfel vs Structured Output Benchmark
Which one should you ship with? Here is the side-by-side panel verdict, pricing read, reviewer split, and community vote comparison.
Developer Tools
Apfel
Tap the free AI already built into your Mac
75%
Panel ship
—
Community
Free
Entry
Apfel is a Swift 6.3 command-line tool that cracks open the on-device language model Apple ships with every Apple Silicon Mac running macOS 26 (Tahoe). Instead of requiring a Claude, OpenAI, or Gemini subscription, Apfel routes through Apple's FoundationModels framework and gives you three interfaces from a single brew install: a pipe-friendly CLI, an interactive chat with context management, and an OpenAI-compatible local HTTP server built on Hummingbird. Under the hood, every token is generated on your Neural Engine and GPU — nothing leaves your machine. The model is roughly 3B parameters with a 4,096-token context window, fast enough for scripting, summarisation, and quick Q&A without latency you'd notice. Pipe-friendly stdin/stdout, JSON output mode, and proper exit codes make it trivially composable with jq, xargs, and shell scripts. The OpenAI-compatible server mode is the killer feature for developers: point any tool that speaks the OpenAI API at localhost and it just works — locally, for free, with zero cold-start. The project is MIT-licensed, started by a solo developer on March 24, 2026, and hit 513 HN points within days of the Show HN post.
Developer Tools
Structured Output Benchmark
The benchmark that tests whether LLMs get JSON values right, not just syntax
75%
Panel ship
—
Community
Free
Entry
Interfaze's Structured Output Benchmark (SOB) exposes a gap that has been quietly breaking production AI pipelines: models can produce syntactically valid JSON while getting the actual values wrong. SOB measures value accuracy across 21 models using 5,000 text passages, 209 OCR documents, and 115 meeting transcripts — scoring each on seven metrics including value accuracy, faithfulness (grounding vs. hallucination), type safety, and perfect-response rate. The benchmark reveals some sobering findings. Even top models like GPT-5.4 and Claude Sonnet 4.6 achieve ~83% on text but drop to 67% on images and only 23.7% on audio. No single model dominates all modalities — GPT-5.4, GLM-4.7, Qwen3.5-35B, and Gemini 2.5 Flash cluster within one point of each other on text. Perfect response rates (all seven metrics correct) rarely exceed 50% for even the best performers. For developers building data extraction pipelines, agents that read invoices, or any system where "correct JSON" means more than syntactically valid JSON, this is required reading. The dataset is on Hugging Face, the paper is on arXiv, and the playground lets you test your own model's structured output capability directly.
Reviewer scorecard
“The OpenAI-compatible server is a genuine unlock — I swapped my local dev config from Ollama to Apfel in two minutes and everything just worked. For Apple Silicon owners who want zero-latency local AI without model downloads, this is the move.”
“This is the benchmark I've been waiting for. 'Valid JSON' is table stakes — the real question is whether field values are correct. This plugs a genuine gap in how we evaluate extraction pipelines.”
“A 3B-parameter model with a 4K context window is impressive for on-device, but it's nowhere near Claude or GPT-5.5 quality. If your task needs real reasoning or long context, you're back to paying for API credits anyway. This is a neat party trick, not a replacement.”
“The 23.7% audio accuracy stat sounds alarming but the test data is text-normalized before scoring, meaning ASR errors are excluded. It's a better benchmark than most but the methodology choices deserve more scrutiny before you rely on it for vendor selection.”
“Apfel is the first glimpse of a world where capable on-device AI comes pre-installed, not downloaded. As Apple's model improves with each macOS release, tools like Apfel will inherit the upgrade for free. The distribution moat Apple is quietly building here is enormous.”
“No universal winner across modalities is the real story here. As agentic systems increasingly handle mixed-media inputs, this exposes that model selection needs to be task-specific. Benchmarks like SOB are how the industry gets smarter about that.”
“I used it to batch-summarise 40 draft posts overnight with a simple shell loop — no API bill, no rate limits, no internet required. For content workflows that need a cheap first pass, it's already practical.”
“For anyone automating content workflows that extract structured data from documents, briefs, or meeting recordings, this tells you which model to actually trust for each media type. Genuinely useful before you commit to an architecture.”
Weekly AI Tool Verdicts
Get the next comparison in your inbox
New AI tools ship daily. We compare them before you waste an afternoon.